Chevrolet's First Independent Front Suspension
One of the most interesting and least reported chapters in Chevrolet history concerns the company's first independent front suspension (IFS). Rather than design their own, Chevrolet purchased the rights to the Dubonnet trailing arm IFS. In some respects, the IFS devised by French racing car driver and auto manufacturer Andre Dubonnet represented the best in European suspension design. Certainly there was nothing like it in this country.
Chevrolet made the Dubonnet "knee action" standard equipment on all 1934 Master passenger cars and sedan delivery vehicles. A modified version with higher spring rates was used on Pontiac Economy Straight Eight cars. After the first year, Chevrolet reverted to the beam axle on top-line Master De Luxe cars, but continued to offer the Dubonnet as an extra-cost option.
The experiment ended in 1939, when the company adopted a more conventional IFS.
Before proceeding, I would like to thank Mike Boteler, VCCA Technical Assistant, for his generous help in researching this article.
Front ends are subject to all sorts of vibration, caused by resonances, wheel imbalance, spring windup during uneven braking and road irregularities. But what doomed the solid front axle was shimmy.
The problem was understood to be a gyroscopic effect caused by angular displacement of the wheels on uneven surfaces. Consider what happens to a beam axle when one front wheel is deflected upward by a road irregularity. The axle rotates in an arc, pushing the wheel on the opposite side down. An angular displacement of a spinning mass, such as a gyroscope or an automobile wheel, generates a torque at right angles to the disturbing force. The operative word is “angular”: a purely vertical motion of the wheel would have no gyroscopic effect.
The driver feels these induced torques as sharp, short-duration kicks on the steering wheel. At high speeds, the kicks come more frequently and the front end imitates a popular dance of the period also known as the shimmy.
Engineers controlled shimmy fairly well when cars rolled on high pressure tires and before front wheels were encumbered with drum brake assemblies. By the late 1920’s it had become clear that no palliative -- neither steering dampeners, irreversible worm steering gears, nor flexible tie rods -- really worked. The only practical solution was to allow the front wheels to move independently of each other.
IFS brought many benefits. Because gyroscopic forces were tamed, the front springs could be as soft (relative to the weight they support) as the rear springs. Suspensions became more compliant for a quantum leap in ride quality. Eliminating the heavy beam axle reduced unsprung weight for better handling over rough surfaces and improved passenger comfort. In addition, IFS freed up needed space around the engine, especially since these new systems usually employed coil, rather than leaf, springs.
The trade off was complexity, fragility (which is why heavy trucks still use beam front axles) and increased body roll. It must have been disconcerting for early customers to feel their cars dive into turns. In addition, IFS forced engineers to increase chassis rigidity to keep the front wheel movement in the vertical, or near vertical, plane. There is no point in designing an elaborate system of wheel restraints, if the supporting platform is allowed to twist.
With the exception of Henry Ford, who loathed the concept, most major US manufacturers went to IFS in 1934. Chrysler and Maurice Olley at Cadillac defined the technology with what came to be known as the SALA (short arm, long arm) suspension. The path of each front wheel was determined by a trapezoid made up two unequal length control arms. Camber -- the tilt of the wheels as viewed from the front of the car -- varied as the suspension worked. This compromised handling, since tires develop maximum adhesion when vertical. But caster -- trail of the wheel behind the king pin --remained constant. Engineers of the time considered caster to be the most important angle, since the righting force developed by the trail of the wheel helps keep the car tracking in a straight line.
All GM divisions, except Chevrolet and Pontiac, adopted versions of Olley’s suspension in 1934. I am not certain why the less expensive cars were left to fend for themselves, but believe it had to do with Sloan’s philosophy of calibrated value for the money. A man might not be so eager to step up to a Buick if a Chevrolet gave the same level of ride sophistication.
There was nothing simple, then or now, about IFS. Olley and his crew worked three years to develop their system, done with the aid of instrumented models and test vehicles. Nor was knowledge about IFS well disseminated outside of the companies. I say that because a computer search showed that the SAE published only four papers between 1928 and 1940 with the term “Independent Front Suspension” in their titles. Chevrolet engineers might not have looked forward to designing a new IFS.
At any rate, Chevrolet chief engineer J.W. Crawford seemed relieved to find an off-the-shelf solution. In 1934 he wrote:
“American manufacturers were fortunate, in that, when they decided on independent front wheel suspension as the best means of obtaining further improvement in riding qualities, the system was already long past the experimental stage. Independent suspension was developed years ago in Europe, where the need was felt earlier than here, and is now in use in some of the foremost British and Continental automobiles... No European car that has adopted front wheel suspension has ever reverted to the old system...”
Speaking at the SAE International Automotive Engineering Congress in Chicago in 1933, Andre Dubonnet made what amounted to a sales pitch. His system was rational in the sense that “front wheels must move parallel to themselves, a condition necessitated by the fact that they have two functional movements; steering action about a vertical pivot and vertical suspension motion.” He also listed specific features, such as 3 1/2 inch of spring deflection, 1 1/2” of free undampened spring deflection for small oscillations, a progressive spring rate, and improved sensitivity because of the absence of shackle and leaf-spring friction.
In 1935 Alfa Romeo fitted Dubonnet pods to their Gran Prix cars and to the new 6c 2300A Pescara sports car. No recommendation could have been higher, although by that time Chevrolet must have been receiving the first field reports about problems with the suspension.
How It Works
The Dubonnet system consists of a pair of oil-filled shock and spring units, hung off the king pins and connected to the wheels by short trailing arms. A transverse link pivots on threaded bushings at the brake backing plates to provide lateral wheel support. Each unit contains two shock absorbers and two coil springs, one inside the other. As the trailing arm rises under load, it cranks the lower spring seat upward, compressing the outer, main spring against the upper spring seat. After 1 7/8” of movement, the secondary spring comes into play to give a progressive spring rate. Preload and ride height can be adjusted externally. The upper shock functions during spring compression, the lower shock dampens rebound. A massive cross member provides the torsional rigidity needed to keep the wheels parallel.
A unique feature is the way all traces of bump steer are eliminated. The tie rod ends mount to the shock units, which pivot on the king pins, but do not move vertically with the wheels. Thus, wheel joust and rebound cannot translate as steering inputs. All other commercial IFS systems which I am aware of require the tie rods to rise and fall as the suspension works.. The tie rods can only be in a geometrically neutral position when the wheels point straight ahead under moderate load.
Faults and Criticisms
In common with Porche and other trailing arm systems, the Dubonnet lacks any provision for anti-dive. In fact, the forward location of the trailing-arm pivots exacerbates the tendency of the vehicle to squat during deceleration. Another, less critical, point is the offset between the king pin centerline and the center of the tire patch contact area. Broulhiet (Ref. 3) reports that the offset was 100 mm, which is large and would generate fairly severe steering inputs in event of a blowout.
But these difficulties were as nothing compared to the oil leaks that developed at the packing glands where the trailing arms entered the shock units. Chevrolet warned owners to check the oil level every 1000 miles (Pontiac said every 2000), but few did. When the shocks ran dry, the totally undampened suspension could bounce for blocks like a proto-low rider. Mike Boteler says some frustrated owners filled the shock units with cup grease, thus eliminating the leaks and freezing the suspension solid. Others retrofitted beam axles. The company went so far as to offer a conversion kit for 1937 - 38 GA and HA models.
Chevrolet did some work on the trailing-arm seals after the first year of production. Flats were added to the packing bolt for wrench purchase, perhaps in an attempt to encourage people to tighten them. But the leak problem was never satisfactorily addressed.
The short arm, long arm IFS that replaced the Dubonnet in 1939 was less ambitious in specification, but more appropriate for the American market.
1. Maurice Olley, “Independent Wheel Suspension -- Its Whys and Wherefores,” SAE paper # 340080, March, 1934.
2. Dyke’s Automobile Encyclopedia, Chicago, 1951, Addenda, p. 35.
3. Andre Dubonnet, “Discussion” appended to Geoge Broulhiet, “Independent Wheel Suspension,” SAE paper # 330029, Sept., 1933.